A system for monitoring the condition of parts of a vehicle

Abstract

A system for monitoring the condition of parts of a vehicle (100) comprises: - battery-operated wireless sensor devices (121-124, 131-134; 201-204; 301- 304) near the monitored parts; - a central node (120, 130; 200; 300) configured to obtain measurements from the sensor devices (121-124, 131-134; 201-204; 301-304); and - a wireless star-topology network coupling the sensor devices (121-124, 131- 134; 201-204; 301-304) to the central node (120, 130; 200; 300). The central node (120, 130; 200; 300) is configured to obtain journey and / or state information for the vehicle (100) and to set a wake-up pattern of one or more of the sensor devices (121-124, 131-134; 201-204; 301-304) based on the journey and / or state information.

Description

A SYSTEM FOR MONITORING THE CONDITION OF PARTS OF A VEHICLEField of the Invention

[0001] The present invention generally relates to monitoring of the condition of maintenance-sensitive parts like bearings, wheels, gearboxes, motors, ... of a vehicle, for instance a train, a tram, a train bogie, an undercarriage or other vehicle riding on tracks. The invention in particular concerns the energy management of a wireless monitoring system comprising battery-operated sensor devices.Background of the Invention

[0002] In order to monitor the condition of maintenance-sensitive parts like bearings, wheels, motors, gearboxes, ... on a vehicle or moving platform, for example a train bogie, the vehicle is typically equipped with a monitoring system comprising a central node, sometimes also called gateway, and sensor devices that communicate with the central node, typically over a star-shaped communication network individually linking each sensor device to the central node.

[0003] Wired monitoring systems exist wherein the central node and sensor devices are wireline powered and the sensor devices communicate over wired connections with the central node.

[0004] Such wired systems have a high installation cost, and the cabling introduces an extra risk for damage during the lifetime. Moreover, wired systems are complex to install in refurbished vehicles, for example train bogies that were initially not equipped with a monitoring system but get a monitoring system as part of their refurbishment.

[0005] On the other hand, wireless monitoring systems exist, comprising battery- operated sensor devices that communicate wirelessly with the central node. Whereas these wireless monitoring systems are less complex to install or integrate in refurbished vehicles, their main challenge is reducing the power consumption of the sensor devices to extend the battery lifetime to several years, preferably over 10 years,thereby avoiding frequent maintenance operations to replace batteries of sensor devices. A well-known technique to extend the lifetime of batteries in sensor devices is polling. A polling-based wireless sensor device wakes-up after a predefined sleep time interval and polls the central node to request if a measurement is desired. If the central node requests a measurement, the sensor device performs the measurement and transmits the measurement result to the central node.

[0006] Polling with a short sleep time results in high power consumption, draining the sensor battery. Polling with a long sleep time results in high latency (time between request for a measurement and the response) and in a high percentage of measurements being carried out when the vehicle is not in nominal operation, for example a train bogie standing still, accelerating, decelerating, riding on a bending track, etc. Measurements however are best taken when the train or vehicle is at nominal operation and under constant environmental conditions: nominal speed, straight track, ...

[0007] United States Patent Application US 2017 / 0210401 , entitled "Railway Vehicle Operations Monitoring", describes a railway vehicle (10) equipped with sensor nodes (14) and a central node named vehicle node (12) to evaluate operating conditions of parts, for example a wheel bearing temperature or wheel vibration data, and to alert situations that require action. Each sensor node is configured to acquire data and communicate with the vehicle node when certain conditions are met. The communication between sensor nodes and vehicle node is wireless. The problem of power consumption of the battery-operated sensor nodes is addressed, and several solutions are proposed to conserve power or limit power consumption. A sensor node for instance enters an active state from a sleep state and returns to the sleep state after a sensor alert is sent to the vehicle node. The sensor nodes further may have a power harvesting component that harvests power from an energy source, for example vibrations, heat, solar, for re-charging the battery. It is recognised that power harvesting might be difficult or impossible (e.g., in an undercarriage). Paragraph

[0063] of US 2017 / 0210401 explains the operation of a sensor node with power harvesting component. In action, the sensor node is in the sleep state where it consumes power sufficient to identify a condition or event resulting in the sensor node waking up, for instance expiry of a timer, or a switch being opened as a result of a shock. Paragraph

[0065] further explains that after waking up, the sensor node performs a power check. If the sensor node has insufficient power to perform an active cycle (including transmission of data to the vehicle node), the sensor returns to the sleep state to accumulate additional power through harvesting. If sufficient power is available, a measurement will be performed. In other words, to obtain a sensor node that spans years of service, performing measurements is balanced against harvesting power.

[0008] In US 2017 / 0210401 , the central vehicle node (12) collects sensor data and possibly also information from external sources such as information regarding the railway vehicle or its cargo from a railroad information system, location information, operating state information of the rail vehicle (e.g., speed) or the like. This is explained in paragraphs

[0079] and

[0081] , The collected data are stored in the vehicle node and used to evaluate whether an alert should be transmitted to a management system.

[0009] The monitoring system known from US 2017 / 0210401 is disadvantageous for several reasons. Firstly, it relies on rather complex sensor nodes with power harvesting technology and intelligence to evaluate their own power consumption in order to autonomously decide between performing an activity or harvesting power in a sleep state. Secondly, the sensor nodes still consume a substantial amount of power in the sleep state because the sleep state requires acquiring data (listening) and event monitoring to decide when to wake up. As a consequence of the ongoing power consumption during the sleep state, a lifetime of several years can only be achieved by integrating power harvesting technologies in the sensor nodes. Yet another drawback of the system known from US 2017 / 0210401 is that most measurements will be taken when the vehicle is not in nominal operation or not under constant environmental condition. As the timing of measurements is determined by events such as a timer, a shock, ... and further by the availability of power, it is likely that the vehicle will be in non-nominal conditions or non-constant environmental conditions, for instance accelerating, decelerating, standing still, riding on a bended track, ... Such measurements under non-nominal conditions or non-constant environmental conditions are less useful to monitor the condition of parts and therefore rather a waste of available battery power.

[0010] United States Patent Application US 2016 / 0152250 entitled "Condition Monitoring System and Method for Monitoring a Condition of a Bearing Unit for a Vehicle" describes a system that monitors the condition of bearings in wheels of a train or truck. The system comprises a central control unit (18 in Fig. 1 ) and battery-operated condition monitoring units or CMUs (10 in Fig. 1 ) each comprising a controller (14 in Fig. 2), transmitter or radio (16 in Fig. 2), and sensor. The control unit (18) is configured to activate / deactivate the CMUs upon reaching predetermined waypoints stored in a waypoint memory. These waypoints are predetermined in view of journey information like long and straight section tracks. As described in

[0037] of US 2016 / 0152250, the control unit incorporates GPS-based location tracking technology to monitor the train's position. The train's GPS coordinates are compared to coordinates of predetermined waypoints programmed in the control unit. When the GPS coordinates of the train coincide with the coordinates of a preprogrammed waypoint, the control unit issues a command to wake the CMUs from a sleep state and trigger data measurements by the sensors.

[0011] In the sleep state described in US 2016 / 0152250, only the data measurements by the sensors are deactivated. The CMU's radio however must remain active to listen for commands issued by the control unit. Such always-on radio drains the battery of the CMU continuously, even when no measurements are performed by the sensor, hence negatively impacts the lifetime of the entire CMU.

[0012] It is an object of the present invention to disclose a wireless monitoring system for the condition of parts of a vehicle that resolves or mitigates one or several of the above-described shortcomings of existing systems. More particularly, it is an object of the invention to provide a scalable monitoring system with reduced installation complexity in new or refurbished vehicles, and / or increased lifetime for sensors, and / or reduced overhead to ensure useful measurements executed when the vehicle is in nominal operations.Summary of the Invention

[0013] According to embodiments of the present invention, the above-defined object is achieved by the system for monitoring the condition of parts of a vehicle as defined by claim 1 , the system comprising:- a plurality of battery-operated wireless sensor devices near the parts monitored;- a central node configured to obtain measurements from the sensor devices;- a wireless star-topology network coupling the sensor devices to the central node, wherein the central node is configured to obtain journey and / or state information wherein journey information comprises information indicative for the current or future trajectory of the vehicle and wherein state information comprises information indicative for the current state of the vehicle, and wherein the central node is further configured to set a wake-up pattern of one or more of the sensor devices based on the journey and / or state information, wherein said wake-up pattern defines one or more times whereon a sensor device has to become active from a deep sleep state wherein said sensor device does not listen for events and does not collect or process data to evaluate if events take place requiring said sensor to become active.

[0014] Thus, according to the present invention, a monitoring system comprises a central node that obtains information on the journey of the vehicle and / or information on the state of the vehicle, and that uses that information to set the wake-up pattern of one or plural sensors. It is noted that in addition to journey and / or state information other types of information may be exploited, like for instance historical measurements. If a past measurement for example exhibited a defect, that information can be used to reduce the time interval in between measurements. If many past measurements on the other hand are not exhibiting any defects, the time interval between measurements can be increased, hence reducing the energy consumption by the battery-operated sensors.

[0015] Controlling the wake-up pattern of a sensor based on journey information and / or state information of the vehicle allows to deprive the sensor of any responsibility for listening for events (sensory or communicative events) during the sleep state and / or evaluating when to wake-up beyond simple evaluation of a clock (e.g. a countdownclock). As a consequence, sensor devices can be implemented with a deep sleep state wherein the power consumption is further reduced, contributing to the invention's aim to extend the lifetime of battery-operated sensor devices in a monitoring system. In such deep sleep state, no external data must be collected or processed to evaluate if events take place requiring the sensor to become active. Thanks to the wake-up pattern set by the central node based on the available journey and / or state information, the times whereon the sensor has to become active are defined and need no further monitoring or evaluation of any data. Only having a clock active on the sensor node to wake-up the sensor device (e.g. the radio thereof) at (a) time(s) specified in the wakeup pattern is sufficient.

[0016] Energy consumption for communication via a radio typically is a factor 1000 higher than energy consumption by a microcontroller and a factor 100 higher than energy consumption by a sensor. It is therefore important to deactivate the radio as much as possible. If one can be triggered by an external device at any time to perform a measurement, the radio must be always-on. Such always-on radio may be implemented as a wake-up radio, a special radio that only reacts on a received signal, but even such wake-up radio consumes an amount of energy that approximately equals the energy consumption of an always-on sensor. Alternatively, the radio may be activated at a regular pace to poll the central unit if a measurement must be performed. The current invention however allows to install dynamic deep sleep states by setting a wake-up pattern based on journey and / or state information. In the deep sleep state, the radio no longer has to listen for signals and the microcontroller no longer has to process data to detect events that trigger polling. In the deep sleep state, the radio is deactivated completely. Only a real-time clock (RTC-clock) is needed to monitor time. Such RTC-clock consumes extremely low energy as a result of which the sensor device is virtually switched off (no energy consumption).

[0017] Journey information in this context encompasses any information indicative for the current or future trajectory the vehicle is travelling, like for instance the condition of rail segments, location of rail segments that are straight and rail segments that are curved, location of stations where the vehicle stands still, location and length of rail segments where a vehicle accelerates, decelerates, rides at nominal speed, etc. for a railway vehicle. Journey information may comprise similar type of information for roadsegments if the vehicle is no railway vehicle. State information in this context encompasses any information indicative for the current state of the vehicle, such as for instance the position or speed of the vehicle, the distance travelled, its load or cargo, the temperature, the time, the time elapsed since the last measurement, the measurement history of one or more sensors, the battery level of one or more sensors, health information of certain parts. Energy management parameter in this context encompasses any configurable parameter that substantially impacts the battery power consumption by the sensor, like for instance the sleep time or sleep pattern of a sensor, the basis of a measurement (frequency-based, time-based, ...), the type of measurement (temperature, vibration, acoustic noise, ...), the duration, sample frequency, bandwidth, filter setting(s), threshold value(s) used for a measurement, which data are communicated to the central node, or the like.

[0018] Key to the present invention is the insight that information on the vehicle's journey and state available at the central node of a parts monitoring system, typically powered by the train or vehicle, can be used to control the setting of sensor device parameters that substantially prolong the lifetime of the sensor battery, typically a battery with low self-discharge like for instance lithium thionyl chloride battery with a self-discharge in the range of 1 % per year, while at the same time guaranteeing that sensor measurements are made at optimal times, i.e. when the vehicle is riding at nominal conditions such that the measurement is not disturbed. The invention this way avoids or mitigates usage of power for useless measurements at non-optimal times. The invention further avoids that complex technologies such as power harvesting technologies must be integrated and managed, while allowing the monitoring system to be wireless such that the monitoring system can be installed easily, even in refurbished vehicles.

[0019] In embodiments of the system according to the present invention, as defined by claim 2, the battery-operated wireless sensor devices do not comprise a power harvesting component.

[0020] Indeed, preferably, the battery of the sensor devices is not rechargeable and no power harvesting and / or battery recharging technologies are integrated. This contributes to the aim of the invention to keep complexity and cost of the system low.A sensor battery lifetime of several years, preferably more than 10 years, is achieved by reducing the amount of measurements and controlling the measurements to take place at optimal times and / or with optimal measurement settings rather than by recharging the batteries as is done in prior art systems.

[0021] In embodiments of the system according to the present invention, as defined by claim 3, the wake-up pattern is different for different sensor devices or different groups of sensor devices.

[0022] Indeed, the wake-up times may be set individually per sensor, for instance based on the available energy left in the battery of the sensor to equalise the lifetime of batteries in multiple sensors thereby elongating the time to the next maintenance of the vehicle. Alternatively, the wake-up pattern or other energy management parameters may be set differently for different groups of sensor devices. The sensor devices may for instance be grouped per type of sensor (accelerometers may have a wake-up pattern that differs from location sensors, acoustic noise sensors, temperature sensors, vibration sensors, etc.) or per type of part that is monitored (suitable times for monitoring the operation of bearings may differ from suitable times for monitoring the operation of gearboxes, motors, etc.).

[0023] In embodiments of the system according to the present invention, as defined by claim 4: the wake-up pattern comprises a sleep time interval of a sensor device; and the central node is configured to determine the sleep time interval from the journey and / or state information and to instruct the sensor device to poll the central node after the sleep time interval.

[0024] Thus, preferred embodiments of the monitoring system according to the invention implement polling. Sensors in sleep mode wake-up at a time defined by the central node in view of the journey and / or state information and then poll the central node to learn from the central node if a measurement is desired. This further contributes to the invention's aim to reduce the power usage of sensors. Sensors hence do not waste battery power for useless measurements: even if their wake-uptime is defined by the central node in view of journey and / or state information, the vehicles journey and / or state may have changed as a result of which the wake-up time of the sensor does not correspond to the anticipated optimal time for executing a measurement. Therefore, the sensor first polls the central node if a measurement is desired and a measurement (including transmission of data to the central node) is executed only after the central node has confirmed that a measurement must be executed. Polling further brings the advantage that synchronisation issues between sensor device and central node are overcome. When polling is used, the exact timing to wake up is less critical. Polling also avoids sensor devices answering at the exact same time. Further, when multiple sensors need to wake-up simultaneously, the central node can only address them one-by-one, so the sensor devices have to continue polling until they get an answer.

[0025] In embodiments of the system according to the invention, as defined by claim 5, the sleep time interval is determined such that the sensor device wakes-up at a time the vehicle is predicted to operate at opportune conditions for a measurement, for instance under nominal conditions and / or under substantially constant environmental parameters.

[0026] Thus, the central node in embodiments of the invention exploits the journey and / or state information to predict when the vehicle will be operating at opportune conditions for a measurement. Opportune conditions may for instance be achieved when the vehicle is under nominal conditions and / or under substantially constant environmental parameters, for instance when a railway vehicle is riding on a straight rail segment at constant speed. This prediction is then used to set the sleep time interval of one or more sensors, such that these sensors do not consume battery power until the point in time is reached where the vehicle is predicted to operate under nominal conditions and / or constant environmental parameters. In order to measure track issues, the measurement may be performed when the vehicle is riding at less nominal conditions but at the time the vehicle is riding at the spot where the track issue is expected.

[0027] In embodiments of the system according to the invention, as defined by claim 6, the sensor device is configured to wake-up after the sleep time interval for polling the central node thereby requesting the central node if a measurement is desired.

[0028] Indeed, even if the sleep time interval is set such that the sensor wakes-up at a point in time where the vehicle is predicted to operate under nominal conditions and / or constant environmental parameters, it remains useful for the sensor to poll the central node if a measurement is desired before executing the measurement. Polling hence remains recommended and further contributes to battery energy savings because it avoids that unnecessary measurements and data transmissions are performed in circumstances where the vehicle is not operating under nominal conditions and / or constant environmental parameters when waking up. This may for instance be the case if a railway vehicle has been standing still longer than expected in a station, if delays are encountered because of construction works, accidents, or other unexpected events have occurred that were not considered when setting the sleep time from the available journey and / or state information.

[0029] In embodiments of the system according to the present invention, as defined by claim 7, the central node is configured to, in response to the polling by the sensor device, either:- instruct the sensor device to perform a measurement;- set a time interval for the sensor device to perform a measurement; or- determine from the journey and / or state information a new sleep time interval for the sensor device.

[0030] Indeed, when being polled, the central node in embodiments of the present invention can use the most actual journey and / or state information of the vehicle and determine if the conditions are optimal to perform a measurement. If this is the case, the central node responds by instructing the sensor to perform a measurement and possibly to transmit the measurement data to the central node. In case the central node establishes that the vehicle is not operating in optimal conditions, for instance due to a delay in its journey or schedule, the central node may determine from the most actual journey and / or state information a new point in time where a measurement would be desired. If this new point in time is in the near future, for instance within a predefinedthreshold of a few minutes, the central node may issue instructions for the sensor to execute the measurement at the anticipated new optimal point in time. The sensor shall use its battery power to perform a useful measurement in a few minutes instead of using the battery power to perform a useless measurement. If however the new point in time lies farther away in time, for instance after the predefined threshold of a few minutes, the central node may set a new sleep time interval for the sensor. The sensor hence shall not perform a useless measurement but returns to the sleep state, further saving battery power.

[0031] In embodiments of the system according to the present invention, as defined by claim 8, the sensor device is further configured to, if the polling remains unanswered by the central node, repeat the polling with progressively increasing in-between time intervals.

[0032] Thus, a sensor that polls the central node but receives no answer from the central node within a predefined time interval, may assume that the central node is asleep and consequently can return to its sleep state as well. Such sensor may wakeup at a later time and poll the central node again. As polling also consumes battery power, it is recommended that the sensor wakes-up and polls the central node with progressively increasing sleep time intervals in between. This will further contribute to increasing the lifetime of the sensor battery.

[0033] In embodiments of the system according to the present invention, as defined by claim 9, the journey and / or state information comprises one or more of:- journey information for the vehicle comprising one or more of:■ a trajectory;■ conditions of rail segments;■ rail segments that are straight;■ rail segments that are curved;■ rail segments where the vehicle accelerates;■ rail segments where the vehicle rides at substantially constant speed;■ rail segments where the vehicle temporally stands still, and / or- state information of the vehicle comprising one or more of:■ position of the vehicle;■ speed of the vehicle;■ distance travelled by the vehicle;■ load information for the vehicle;■ temperature information;■ time information;■ time elapsed since the last measurement by a sensor device;■ measurement history information of the sensor devices;■ battery level information for one or more of the sensor devices;■ health information for hardware in the vehicle.

[0034] Although many examples are given of journey information and state information that could be useful to set an energy management parameter of one or more sensors, the skilled person will appreciate that the above list of examples is non-exhaustive and other journey information and / or state information could be exploited by the central node to set an energy management parameter value of one or more sensors.

[0035] In embodiments of the system according to the invention, as defined by claim 10, the central node is configured to receive the journey and / or state information at least partially from a Passenger Information System aboard the vehicle.

[0036] Indeed, when connected to a Passenger Information System, the central node of the monitoring system may exploit the journey and / or state information typically available in such Passenger Information System to set one or more energy management parameter, for instance the sleep pattern or sleep time of a sensor. Such Passenger Information System is typically present aboard a railway vehicle such as a train, tram or metro, but is nowadays only used to inform passengers on the journey or state of the vehicle, e.g. delays in the schedule. Exploiting this information for setting energy management parameters of wireless battery-operated sensors thereby controlling when measurements are made and transmitted to the central node of a monitoring system will significantly contribute to reducing the operational costs of the vehicle as downtime for replacing batteries will be minimised and data collected through measurements will be more useful and therefore more reliable to estimate the condition of maintenance-sensitive parts like bearings, gear boxes, wheels and motors.

[0037] In embodiments of the system according to the present invention, as defined by claim 11 , the central node is configured to receive the journey and / or state information at least partially from a Train Control Management System, abbreviated TCMS, aboard the vehicle.

[0038] Indeed, when connected to a TCMS, the central node of the monitoring system may exploit the journey and / or state information typically available in such TCMS to set one or more energy management parameter, for instance the sleep pattern or sleep time of a sensor. Such TCMS is typically present aboard a railway vehicle such as a train, tram or metro.

[0039] Alternatively, or supplementary, as defined by claim 12, the central node in embodiments of the system according to the invention comprises machine learning technology adapted to generate at least part of the journey and / or state information through automated learning.

[0040] In particular railway vehicles such as trains, trams or metros, but also other vehicles used in public or private transport such as busses or trucks, are often used to travel the same trajectory periodically, for example daily or weekly. For such vehicles, it will be advantageous to have machine learning technology aboard that automatically learns the journey and / or state of the vehicle and / or track at respective times of the day or week. Such journey and / or state information obtained through automated learning may be exploited by a central node, either alone or in combination with journey and / or state information obtained from other systems, to intelligently set energy management parameters of sensors of a monitoring system aboard the vehicle.

[0041] In embodiments of the system according to the present invention, as defined by claim 13, wherein the central node is configured to set an energy management parameter of one or more of the sensor devices based on the journey and / or state information, the energy management parameter comprising one or more of:- basis of a measurement by a sensor device, being either a frequency-based measurement or a time-based measurement;- type of measurement by a sensor device, being either temperature, vibration, acoustic noise;- duration of a measurement by a sensor device;- sample frequency used by a sensor device for a measurement;- bandwidth used by a sensor device for a measurement;- filter settings used by a sensor device for a measurement;- a threshold value used by a sensor for a measurement;- information to be communicated by a sensor device for a measurement to the central node.

[0042] Although many examples are given of energy management parameters of sensor devices that could be set intelligently by the central node using journey and / or state information, the skilled person will appreciate that the above list of examples is non-exhaustive and other energy management parameters of one or more sensors could be set likewise by the central node.Brief Description of the Drawings

[0043] Fig. 1 shows a train comprising a plurality of bogies equipped with embodiments of the parts monitoring system according to the present invention;

[0044] Fig. 2 shows a single bogie equipped with an embodiment of the parts monitoring system according to the present invention;

[0045] Fig. 3 is a functional block scheme of an embodiment of the parts monitoring system according to the present invention;

[0046] Fig. 4 is a time diagram illustrating the communication between the central node and a sensor device in an embodiment of the parts monitoring system according to the present invention; and

[0047] Fig. 5 shows an example embodiment of a suitable computing system 500 for performing one or several steps in embodiments of the parts monitoring system according to the invention.Detailed Description of Embodiment(s)

[0048] Fig. 1 shows a train 100 comprising a locomotive 100 and railway carriages 102, 103. The railway carriages 102, 103 each comprise a bogie equipped with respective embodiments of the parts monitoring system according to the invention. The bogie of carriage 102 is equipped with a central node 120 and four battery-operated sensor devices 121 , 122, 123, 124 that communicate with the central node 120 through a startopology wireless network. The sensor devices 121 , 122, 123 and 124 are mounted on respective wheels of carriage 102 and for instance comprise accelerometers used to measure degrade over time such as defects in the inner raceway of bearings, defects in the outer raceway of bearings, defects in rollers of bearings. Similarly, the bogie of carriage 103 is equipped with a central node 130 and four battery-operated sensor devices 131 , 132, 133, 134 that communicate with the central node 130 through a startopology wireless network. The sensor devices 131 , 132, 133 and 134 are mounted on respective wheels of carriage 103. It is noted that the drawn configuration with starshaped wireless networks per bogie is preferred for passenger trains. In freight trains wherein for example only the locomotive is electrically powered, a long-range starshaped wireless network or alternatively a wireless mesh network could be deployed connecting the sensor devices mounted on plural carriages or bogies to a single central node installed in the locomotive such that not every carriage or bogie has to be equipped with a central node that requires electrical powering.

[0049] Fig. 2 shows a single train bogie equipped with an embodiment of the parts monitoring system according to the present invention. More particularly, Fig. 2 shows a bogie frame 210 and two wheelsets 220, 230. The first wheelset 220 comprises an axle 221 and two wheels 222, 223. The axle 221 and wheels 222, 223 rotate via wheel bearings 224, 225 mounted on the bogie frame 210. The second wheelset 230 also comprises an axle 231 and two wheels 232, 233. The axle 231 and wheels 232, 233 rotate via wheel bearings 234, 235 mounted on the bogie frame 210. The parts monitoring system comprises a central node 200 and four battery-operated sensor devices 201 , 202, 203 and 204, each mounted on or near a respective wheel. The battery-operated sensor devices 201 , 202, 203 and 204 are connected to the central node 200 through a wireless star-topology network. Each of the sensor devices 201 -204 hence is equipped with a battery and a wireless transceiver, for instance a Bluetooth transceiver or WiFi transceiver. Depending on its type, each sensor device further comprises additional components. In the example if Fig. 2 for instance, the sensor devices 201-204 comprise an accelerometer. Alternatively, one or more of the sensor devices could comprise a thermometer.

[0050] Fig. 3 is a functional block scheme of an embodiment of the parts monitoring system according to the present invention. The essential components of such parts monitoring system are a central node 300, typically electrically powered through wire powering or the vehicle's battery, and plural sensor devices 301 , 302, 303, 304, battery powered by a sensor battery in or near the sensor device, and wirelessly connected to the central node 300.

[0051] Fig. 4 is a time diagram illustrating the communication between the central node300 and a sensor device 301 in an embodiment of the parts monitoring system like the one drawn in Fig. 3.

[0052] During phase 411 the central node 300 obtains information on the journey of the train and / or information on the state of the train and predicts from the journey and / or state information when the train will be operating at opportune conditions for sensor301 to perform a measurement. The journey and / or state information may be obtained from a Passenger Information System (PIS), from a Train Control Management System (TCMS), from another external source, through automated learning, or through the combination of a subset or all of the foregoing. Journey information may comprise information on the upcoming trajectory of the train like the start location or station and start time, the destination location or station and estimated time of arrival, in-between stations and estimated times of arrival there, information on conditions of rail segments, information indicating that rail segments are straight or curved, information indicating on which rail segments the train typically accelerates, typically decelerates, typically rides at substantially constant speed or temporally stands still, information on maintenance or construction works ongoing at certain rail segments, etc. The journey information may be obtained before the start of a journey or during the journey. Obtained journey information may be updated during the journey as well. State information may comprise information on the actual position of the train, the actualspeed of the train, the distance already travelled by the train, load information of the train, temperature information, time information, time elapsed since the last measurement by a sensor device, measurement history information of the sensor devices, battery level information for one or more of the sensor devices, health information for hardware in train, etc. The state information typically is obtained from various sensors present in the train or from a database collecting and memorizing information from sensors. The journey information and / or state information enable the central node 300 to calculate points in time where it will be opportune for sensor 301 to perform a measurement. For instance, nominal conditions or substantially constant environmental parameters, like for example the train riding at constant pace on a straight rail segment for which no track issues are registered, could be an opportune time for an accelerometer in sensor 301 to perform a useful measurement to establish the conditions of bearings. Alternatively, to measure track issues on a particular rail segment, the train may be operating at non-nominal conditions but it is important that the sensor measurement takes place at the instant the train rides on the spot or rail segment where the track issue is expected. The skilled person will appreciate that numerous other opportune times for sensor measurements can be defined depending on what part(s) is / are monitored and that it is possible to configure a central node 300 with processing capabilities to calculate such opportune times from journey information and / or state information that is obtained in advance or real-time. At the end of phase 411 , the central node 300 hence is aware of an opportune time or multiple opportune times for sensor 301 to perform a measurement.

[0053] In the embodiment of Fig. 4, the central node 300 calculates a sleep time for sensor 301 from the actual time and the opportune time for measurement. The sleep time is for instance calculated such that the sensor wakes up 2 minutes ahead of the opportune time for measurement. In Fig. 4, 412 represents a message or signal sent from the central node 300 to the sensor 301 to inform the sensor 301 of the sleep time. Sensor 301 , upon receipt of message or signal 412, hence can stay in a deep sleep state 413 for a time interval with length corresponding to the sleep time. During the sleep state 413, no listening for events is necessary and battery power consumption by the sensor 301 is reduced to the utmost minimum. Also, time synchronisation between the central node 300 and sensor 301 is not necessary, further reducing the battery power consumption by the senor 301 .

[0054] When the sleep time is over, sensor 301 initiates polling. The sensor 301 awakes and sends a poll request message or signal 414 to the central node 300. The poll request message or signal 414 is sent to verify if the central node 300 still wants the sensor 301 to perform a measurement. Journey information and / or state information may have changed as a result of which the opportune time for a measurement may have changed.

[0055] Upon receipt of the poll request message or signal 414, the central node 300 verifies if the time is still opportune for a sensor measurement in view of the most recent journey and / or state information, and sends a poll response message or signal 415 to the sensor 301 . In the poll response message or signal 415, the central node 300 either instructs the sensor 301 to perform the measurement, communicates a waiting time interval for the sensor 301 to wait before performing the measurement, or communicates a new sleep time for the sensor 301 to return to the deep sleep state without performing a measurement. The polling procedure wherein the sensor 301 at the end of the sleep time requests the central node 300 to confirm that a measurement must be executed further enhances the battery lifetime of the sensor 301 because it avoids that useless measurements and data transmissions are performed at inopportune times.

[0056] When the sensor 301 receives a poll response message or signal 415 that instructs to perform a measurement, as is the case in Fig. 4, a measurement 416 of metrics is performed by sensor 301 , and the measurement data 417 are transmitted from the sensor 301 to the central node 300. In case no poll response message or signal 415 is received by the sensor 301 within a predetermined time interval from transmitting the poll request message 414, the sensor 301 repeats transmitting the poll request message or signal 414 with time intervals in between subsequent transmissions that are progressively increasing.

[0057] Although a sleep time interval of sensor 301 is calculated from the journey and / or state information in the above example illustrated by Fig. 4, it is noted that the central node 300 could exploit the journey and / or state information in other embodiments of the invention to calculate alternate or additional parameter values forenergy management parameters of one or more battery-operated sensors. Examples of such energy management parameters are the basis for a measurement by a sensor device, i.e. either a frequency-based measurement or a time-based measurement, the type of measurement by a sensor device, i.e. either temperature, vibration, acoustic noise, etc., the duration of a measurement by a sensor device, the sample frequency used by a sensor device for a measurement, the bandwidth used by a sensor device for a measurement, filter settings used by a sensor device for a measurement, a threshold value used by a sensor for a measurement, which information to be communicated by a sensor device for a measurement to the central node.

[0058] Fig. 5 shows a suitable computing system 500 enabling to perform one or several steps in embodiments of the parts monitoring system according to the invention. Computing system 500 may in general be formed as a suitable general- purpose computer and comprise a bus 510, a processor 502, a local memory 504, one or more optional input interfaces 514, one or more optional output interfaces 516, a communication interface 512, a storage element interface 506, and one or more storage elements 508. Bus 510 may comprise one or more conductors that permit communication among the components of the computing system 500. Processor 502 may include any type of conventional processor or microprocessor that interprets and executes programming instructions. Local memory 504 may include a random-access memory (RAM) or another type of dynamic storage device that stores information and instructions for execution by processor 502 and / or a read only memory (ROM) or another type of static storage device that stores static information and instructions for use by processor 502. Input interface 514 may comprise one or more conventional mechanisms that permit an operator or user to input information to the computing device 500, such as a keyboard 520, a mouse 530, a pen, voice recognition and / or biometric mechanisms, a camera, etc. Output interface 516 may comprise one or more conventional mechanisms that output information to the operator or user, such as a display 540, etc. Communication interface 512 may comprise any transceiver-like mechanism such as for example one or more Ethernet interfaces that enables computing system 500 to communicate with other devices and / or systems, for example with other computing devices 581 , 582, 583. The communication interface 512 of computing system 500 may be connected to such another computing system by means of a local area network (LAN) or a wide area network (WAN) such as for example theinternet. Storage element interface 506 may comprise a storage interface such as for example a Serial Advanced Technology Attachment (SATA) interface or a Small Computer System Interface (SCSI) for connecting bus 510 to one or more storage elements 608, such as one or more local disks, for example SATA disk drives, and control the reading and writing of data to and / or from these storage elements 508. Although the storage element(s) 508 above is / are described as a local disk, in general any other suitable computer-readable media such as a removable magnetic disk, optical storage media such as a CD or DVD, -ROM disk, solid state drives, flash memory cards, ... could be used. Computing system 500 could thus correspond to the processing circuitry in the central nodes 120, 130, 200, 300 in the embodiments illustrated by Figs. 1-4.

[0059] Although the present invention has been illustrated by reference to specific embodiments, it will be apparent to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied with various changes and modifications without departing from the scope thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. In other words, it is contemplated to cover any and all modifications, variations or equivalents that fall within the scope of the basic underlying principles and whose essential attributes are claimed in this patent application. It will furthermore be understood by the reader of this patent application that the words "comprising" or "comprise" do not exclude other elements or steps, that the words "a" or "an" do not exclude a plurality, and that a single element, such as a computer system, a processor, or another integrated unit may fulfil the functions of several means recited in the claims. Any reference signs in the claims shall not be construed as limiting the respective claims concerned. The terms "first", "second", third", "a", "b", "c", and the like, when used in the description or in the claims are introduced to distinguish between similar elements or steps and are not necessarily describing a sequential or chronological order. Similarly, the terms "top", "bottom", "over", "under", and the like are introduced for descriptive purposes and not necessarily to denote relative positions. It is to be understood that the terms so used are interchangeableunder appropriate circumstances and embodiments of the invention are capable of operating according to the present invention in other sequences, or in orientations different from the one(s) described or illustrated above.

Claims

CLAIMS1 . A system for monitoring the condition of parts of a vehicle (100), said system comprising:- a plurality of battery-operated wireless sensor devices (121-124, 131 -134; 201 - 204; 301-304) near said parts monitored;- a central node (120, 130; 200; 300) configured to obtain measurements from said sensor devices (121 -124, 131 -134; 201 -204; 301-304);- a wireless star-topology network coupling said sensor devices (121 -124, 131 - 134; 201-204; 301-304) to said central node (120, 130; 200; 300), wherein said central node (120, 130; 200; 300) is configured to obtain journey and / or state information wherein journey information comprises information indicative for the current or future trajectory of said vehicle (100) and wherein state information comprises information indicative for the current state of the vehicle (100),CHARACTERIZED IN THAT said central node (120, 130; 200; 300) is further configured to set a wake-up pattern of one or more of said sensor devices (121 -124, 131 -134; 201-204; 301 -304) based on said journey and / or state information, wherein said wake-up pattern defines one or more times whereon a sensor device has to become active from a deep sleep state wherein said sensor device does not listen for events and does not collect or process data to evaluate if events take place requiring said sensor to become active.

2. The system according to claim 1 , wherein said battery-operated wireless sensor devices (121-124, 131 -134; 201- 204; 301-304) do not comprise a power harvesting component.

3. The system according to claim 1 , wherein said wake-up pattern is different for different sensor devices or different groups of sensor devices.

4. The system according to any of the preceding claims, wherein said wake-up pattern comprises a sleep time interval of a sensor device (301 ); andwherein said central node (300) is configured to determine said sleep time interval (412) from said journey and / or state information and to instruct said sensor device (301 ) to poll said central node after said sleep time interval (412).

5. The system according to claim 4, wherein said sleep time interval (412) is determined such that said sensor device wakes-up at a time said vehicle is predicted to operate under nominal conditions and / or under substantially constant environmental parameters.

6. The system according to claim 4 or claim 5, wherein said sensor device (301 ) is configured to wake-up after said sleep time interval for polling (414) said central node (300) thereby requesting said central node (300) if a measurement is desired.

7. The system according to claim 6, wherein said central node (300) is configured to, in response to said polling (414) by said sensor device (301 ), either:- instruct said sensor device (301 ) to perform a measurement;- set a time interval for said sensor device (301 ) to perform a measurement; or- determine from said journey and / or state information a new sleep time interval for said sensor device (301 ).

8. The system according to claim 4 or claim 5, wherein said sensor device (301 ) is further configured to, if said polling (414) remains unanswered by said central node, repeat said polling with progressively increasing in-between time intervals.

9. The system according to any of the preceding claims, wherein said journey and / or state information comprises one or more of: - journey information for said vehicle (100) comprising one or more of:■ a trajectory;■ conditions of rail segments;■ rail segments that are straight;■ rail segments that are curved;■ rail segments where said vehicle (100) accelerates;■ rail segments where said vehicle (100) rides at substantially constant speed;■ rail segments where said vehicle (100) temporally stands still, and / or- state information of said vehicle (100) comprising one or more of:■ position of said vehicle (100);■ speed of said vehicle (100);■ distance travelled by said vehicle (100);■ load information for said vehicle (100);■ temperature information;■ time information;■ time elapsed since the last measurement by a sensor device;■ measurement history information of said sensor devices (121 -124, 131- 134; 201-204; 301-304);■ battery level information for one or more of said sensor devices (121- 124, 131-134; 201-204; 301 -304);■ health information for hardware in said vehicle (100).

10. The system according to any of the preceding claims, wherein said central node (120, 130; 200; 300) is configured to receive said journey and / or state information at least partially from a Passenger Information System aboard said vehicle (100).11 . The system according to any of the preceding claims, wherein said central node (120, 130; 200; 300) is configured to receive said journey and / or state information at least partially from a Train Control Management System aboard said vehicle (100).

12. The system according to any of the preceding claims, wherein said central node (120, 130; 200; 300) comprises machine learning technology adapted to generate at least part of said journey and / or state information through automated learning.

13. The system according to any of the preceding claims,wherein said central node (120, 130; 200; 300) is configured to set an energy management parameter of one or more of said sensor devices (121-124, 131 -134; 201 -204; 301-304) based on said journey and / or state information, said energy management parameter comprising one or more of: - basis of a measurement by a sensor device, being either a frequency-based measurement or a time-based measurement;- type of measurement by a sensor device, being either temperature, vibration, acoustic noise;- duration of a measurement by a sensor device; - sample frequency used by a sensor device for a measurement;- bandwidth used by a sensor device for a measurement;- filter settings used by a sensor device for a measurement;- a threshold value used by a sensor for a measurement;- information to be communicated by a sensor device for a measurement to said central node.

Images